ON THE PENETRATION OF HEAT ACROSS 
LAYERS OF GAS. 
BY 
G. JOHNSTONE STONEY, m.a., F.x.8., Secretary Royal Dublin Society. 
[Read 21st May, 1877.] 
(A bstract.) 
THE name penetration is suggested in this communication for 
the way in which heat leaks away across the curtailed Crookes’s 
layer which comes into existence when gas intervenes between a 
heater and cooler brought sufficiently close. 
The laws of the escape of heat by penetration are approximately 
traced, and it is shown that the rate of cooling which occurs from 
this cause will often be considerable. 
It is further shown that the laws of the escape of heat by 
penetration completely account for a large body of remarkable 
results obtained upwards of thirty years ago by De la Provostaye 
and Desains when experimenting upon the escape of heat from 
thermometer bulbs placed within exhausted receivers of different 
sizes. They found that when the receiver was small, and when 
the exhaustion passed a certain point, there was a more rapid 
cooling than could be accounted for by Dulong and Petit’s laws 
for radiation and convection. When the exhaustion was continued 
this additional loss of heat rose gradually to a maximum after 
which it rapidly fell off, but still remained of considerable amount 
at the lowest tensions they could command, when the receiver 
was only 6 cm. across. All these results are shown to be in con- 
formity with the deductions from the theory. 
It is also shown that the abnormal law for convection in hydro- 
gen which Dulong and Petit derived from their experiments, is 
accounted for by assuming that the law for convection is the same 
or nearly the same in all gases, but that what these physicists 
supposed to be entirely convection was in reality the mixed effect 
of convection and penetration. 
